1 of 17

Better Findable Norms (= More Use = More Impact)

22-4-2022

Kadaster Data Science Team

Lexi Rowland, Wouter Beek, Erwin Folmer

​

​

Who knows NEN 7522?

​

“Normal users don’t search for NEN 13765,

but search for related norms to fire safety”

​

​

​

If you know the norm number you will find it: skip this talk

If you don’t know the norm number: watch this talk!

Which norm might be relevant for me?

Template indeling

2 of 17

1. The Problem

6 months??!?!?!

What about 6 days???

​

Prototypes?!?!?!

What about near production?

​

Simplicity is key!

Unlock NEN data: Yes we can!

​

For NEN: better findable standards = wider standards adoption = more use = more impact

Or

Not findable standards ≠ wider adoption ≠ more use ≠ more impact!!!

It is fundamental for NEN: (Better) Findable Standards.

2 |

Template indeling

3 of 17

2. Simple Solution:

Standards-based Norm Summaries

Norm Summaries:

  • …are generated for all Norms.
  • …are stored in an open standards based template. (Schema.org)
  • …are published in a RDF triplestore with API’s. (SPARQL, Elastic Search).

​

  • …are fundamental for many search applications. (text search, serendipity, recommender)

3 |

Template indeling

4 of 17

3: The Project (implementation)

Part 1 - Fundamental: Norm Summary generation and publication

  • “Near production ready” - No regret investment

​

Part 2: Search Applications “on top of”

  • So many opportunities - We did:
    • Querying & Text Search (out of the box - “for free”)
    • Serendipity Search (experimental)
    • Recommender System (experimental)

​

4 |

Template indeling

5 of 17

DEMO TIME!

screenshots/links & live

Template indeling

6 of 17

The Demo - Cool, but won’t show live

please have a look yourself!

https://data.labs.kadaster.nl/nen/-/overview

Content:

  1. Norm Summaries (all) published in Triple Store
    1. You can navigate through relations in Norms
    2. You can use the APIs
      1. SPARQL queries
      2. Elastic Search

​

  • Norm Summary format “as open standard”
    • (Norm Summary Model (mapping schema.org))
    • Schema.org validator (Google Metadata Crawler)

6 |

Template indeling

7 of 17

  1. Norm Summaries published in Triplestore

7 |

Template indeling

8 of 17

1a. You can navigate through relations in Norms

8 |

Template indeling

9 of 17

1b. You can use the APIs i. SPARQL

9 |

Template indeling

10 of 17

1b. You can use the APIs ii. Elasticsearch

10 |

Template indeling

11 of 17

2a. Norm Summary Model (mapping schema.org)

11 |

Template indeling

12 of 17

2b. Schema.org validator (Google Metadata Crawler)

12 |

Template indeling

13 of 17

The Demo - Even cooler, will show live

Part Two - Search Applications

​

  1. Serendipity Search (Lexi)
  2. Weaver demo
  3. (Data Story)

​

  • Recommender System (Wouter)
  • Data Story

​

13 |

Template indeling

14 of 17

1. Serendipity Search

Link: https://kadaster.wvr.io/nen-hackathon-22?branch=main&tab=attributes

​

14 |

Template indeling

15 of 17

Backup: Semantic Search

15 |

Template indeling

16 of 17

2. Recommender System

16 |

Template indeling

17 of 17

Better Findable Norms (= More Use = More Impact)

22-4-2022

Kadaster Data Science Team

Lexi Rowland, Wouter Beek, Erwin Folmer

Template indeling